Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces

Gerard Roma; Owen Green; Pierre Alexandre Tremblay

Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces

Gerard Roma, Owen Green, Pierre Alexandre Tremblay

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Descriptor spaces have become an ubiquitous interaction paradigm for music based on collections of audio samples. However, most systems rely on a small predefined set of descriptors, which the user is often required to understand
and choose from. There is no guarantee that the chosen descriptors are relevant for a given collection. In addition,this method does not scale to longer samples that require higher-dimensional descriptions, which biases systems towards the use of short samples. In this paper we propose novel framework for automatic creation of interactive sound spaces from sound collections using feature learning and dimensionality reduction. The framework is implemented as a
software library using the SuperCollider language. We compare several algorithms and describe some example interfaces for interacting with the resulting spaces. Our experiments signal the potential of unsupervised algorithms for creating data-driven musical interfaces.

Original language	English
Title of host publication	Proceedings of the International Conference on New Interfaces for Musical Expression
Editors	Marcelo Queiroz, Anna Xambó Sedó
Place of Publication	Porto Alegre
Pages	313-318
Number of pages	6
Publication status	Published - Jun 2019
Event	The International Conference on New Interfaces for Musical Expression - Porto Alegre, Brazil Duration: 3 Jun 2019 → 6 Jun 2019 https://www.nime.org/

Publication series

Name	Proceedings of the conference on New Interface for Musical Expression (NIME)
ISSN (Print)	2220-4806

Conference

Conference	The International Conference on New Interfaces for Musical Expression
Abbreviated title	NIME2019
Country	Brazil
City	Porto Alegre
Period	3/06/19 → 6/06/19
Internet address	https://www.nime.org/

Access to Document

Fingerprint Dive into the research topics of 'Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces'. Together they form a unique fingerprint.

2 Invited talk

The FluCoMa Project: Empowering techno-fluent sound artists
Pierre Alexandre Tremblay (Speaker)
15 Feb 2020
Activity: Talk or presentation types › Invited talk
Fluid Corpus Manipulation: blurring taxonomies through creative convergences of practices
Pierre Alexandre Tremblay (Speaker)
2 May 2019
Activity: Talk or presentation types › Invited talk

1 Active

FluCoMa: Fluid Corpus Manipulations
Tremblay, P. A., Green, O., Roma, G., Harker, A., Clarke, M. & Dufeu, F.
1/09/17 → 31/08/22
Project: Research

Cite this

Roma, G., Green, O., & Tremblay, P. A. (2019). Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces. In M. Queiroz, & A. X. Sedó (Eds.), Proceedings of the International Conference on New Interfaces for Musical Expression (pp. 313-318). (Proceedings of the conference on New Interface for Musical Expression (NIME)).. https://www.nime.org/proceedings/2019/nime2019_paper060.pdf

Roma, Gerard ; Green, Owen ; Tremblay, Pierre Alexandre. / Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces. Proceedings of the International Conference on New Interfaces for Musical Expression. editor / Marcelo Queiroz ; Anna Xambó Sedó. Porto Alegre, 2019. pp. 313-318 (Proceedings of the conference on New Interface for Musical Expression (NIME)).

@inproceedings{2305974dc3144c26b60d5b83d51cd1a3,

title = "Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces",

abstract = "Descriptor spaces have become an ubiquitous interaction paradigm for music based on collections of audio samples. However, most systems rely on a small predefined set of descriptors, which the user is often required to understandand choose from. There is no guarantee that the chosen descriptors are relevant for a given collection. In addition,this method does not scale to longer samples that require higher-dimensional descriptions, which biases systems towards the use of short samples. In this paper we propose novel framework for automatic creation of interactive sound spaces from sound collections using feature learning and dimensionality reduction. The framework is implemented as asoftware library using the SuperCollider language. We compare several algorithms and describe some example interfaces for interacting with the resulting spaces. Our experiments signal the potential of unsupervised algorithms for creating data-driven musical interfaces.",

keywords = "Dimensionality reduction, feature learning, information visualisation",

author = "Gerard Roma and Owen Green and Tremblay, {Pierre Alexandre}",

year = "2019",

month = jun,

language = "English",

series = "Proceedings of the conference on New Interface for Musical Expression (NIME)",

pages = "313--318",

editor = "Marcelo Queiroz and Sed{\'o}, {Anna Xamb{\'o}}",

booktitle = "Proceedings of the International Conference on New Interfaces for Musical Expression",

note = "The International Conference on New Interfaces for Musical Expression, NIME2019 ; Conference date: 03-06-2019 Through 06-06-2019",

url = "https://www.nime.org/",

}

Roma, G , Green, O & Tremblay, PA 2019, Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces. in M Queiroz & AX Sedó (eds), Proceedings of the International Conference on New Interfaces for Musical Expression. Proceedings of the conference on New Interface for Musical Expression (NIME), Porto Alegre, pp. 313-318, The International Conference on New Interfaces for Musical Expression, Porto Alegre, Brazil, 3/06/19. <https://www.nime.org/proceedings/2019/nime2019_paper060.pdf>

Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces. / Roma, Gerard ; Green, Owen ; Tremblay, Pierre Alexandre.

Proceedings of the International Conference on New Interfaces for Musical Expression. ed. / Marcelo Queiroz; Anna Xambó Sedó. Porto Alegre, 2019. p. 313-318 (Proceedings of the conference on New Interface for Musical Expression (NIME)).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces

AU - Roma, Gerard

AU - Green, Owen

AU - Tremblay, Pierre Alexandre

PY - 2019/6

Y1 - 2019/6

N2 - Descriptor spaces have become an ubiquitous interaction paradigm for music based on collections of audio samples. However, most systems rely on a small predefined set of descriptors, which the user is often required to understandand choose from. There is no guarantee that the chosen descriptors are relevant for a given collection. In addition,this method does not scale to longer samples that require higher-dimensional descriptions, which biases systems towards the use of short samples. In this paper we propose novel framework for automatic creation of interactive sound spaces from sound collections using feature learning and dimensionality reduction. The framework is implemented as asoftware library using the SuperCollider language. We compare several algorithms and describe some example interfaces for interacting with the resulting spaces. Our experiments signal the potential of unsupervised algorithms for creating data-driven musical interfaces.

AB - Descriptor spaces have become an ubiquitous interaction paradigm for music based on collections of audio samples. However, most systems rely on a small predefined set of descriptors, which the user is often required to understandand choose from. There is no guarantee that the chosen descriptors are relevant for a given collection. In addition,this method does not scale to longer samples that require higher-dimensional descriptions, which biases systems towards the use of short samples. In this paper we propose novel framework for automatic creation of interactive sound spaces from sound collections using feature learning and dimensionality reduction. The framework is implemented as asoftware library using the SuperCollider language. We compare several algorithms and describe some example interfaces for interacting with the resulting spaces. Our experiments signal the potential of unsupervised algorithms for creating data-driven musical interfaces.

KW - Dimensionality reduction

KW - feature learning

KW - information visualisation

UR - https://www.nime.org/

M3 - Conference contribution

T3 - Proceedings of the conference on New Interface for Musical Expression (NIME)

SP - 313

EP - 318

BT - Proceedings of the International Conference on New Interfaces for Musical Expression

A2 - Queiroz, Marcelo

A2 - Sedó, Anna Xambó

CY - Porto Alegre

T2 - The International Conference on New Interfaces for Musical Expression

Y2 - 3 June 2019 through 6 June 2019

ER -

Adaptive Mapping of Sound Collections for Data-driven Musical Interfaces

Abstract

Publication series

Conference

Access to Document

The FluCoMa Project: Empowering techno-fluent sound artists

Fluid Corpus Manipulation: blurring taxonomies through creative convergences of practices

FluCoMa: Fluid Corpus Manipulations

Cite this